/srv/irclogs.ubuntu.com/2015/03/03/#ubuntu-ci-eng.txt

imgbot=== IMAGE 119 building (started: 20150303-02:05) ===02:05
=== chihchun_afk is now known as chihchun
Mirvalexabreu: it seems the rtm-013 was rebuilt before the trunk was properly updated, thus it's not publishable and needs another rebuild + testing instead06:51
=== marcusto_ is now known as marcustomlinson
=== marcusto_ is now known as marcustomlinson_
=== marcustomlinson_ is now known as marcustomlinson
popeyMirv: will probably miss the landing meeting as I have an appointment at school.07:48
Mirvpopey: ok07:52
=== marcusto_ is now known as marcustomlinson
ogra_rvr, how did it fail ?08:44
ToyKeeperogra_, sergiusens, robru: Could you take a look at http://pad.lv/1427559 to review a phablet-tools fix needed for automated sanity testing?08:50
ubot5Launchpad bug 1427559 in phablet-tools (Ubuntu) "phablet-network requires NetworkManager, sudo, and wifi" [Undecided,New]08:50
ToyKeeper(patch already linked)08:51
ogra_ToyKeeper, any reason you can not use the existing -n option ?08:52
ogra_(we use it in the lab)08:53
ToyKeeperFor one, it requires patching every tool which uses phablet-network...08:53
ToyKeeper(which expect phablet-network to "just work" with no options or input)08:54
ogra_well, couldn't you just add a default for that file instead then ?08:55
ToyKeeperI suppose it could use an environment variable or dotfile to point toward a NM config file, and eliminate a few lines to parse the other file...08:56
ToyKeeper(and use that instead of '-n', when available)08:57
ogra_ToyKeeper, http://paste.ubuntu.com/10512922/09:00
ogra_err09:00
ogra_http://paste.ubuntu.com/10512925/09:00
ogra_the second one is the right way around :)09:01
ToyKeeperogra_: Are you suggesting it as a patch, or suggesting that each affected person modify /usr/bin/phablet-network?  I'm trying to avoid the need to modify files installed by apt-get.09:04
ogra_huh ? i suggeted to change your MP :) ... but there are bugs ... one sec09:05
jibelogra_, what you propose makes no difference with current behaviour because you always have to provide a network file as argument to -n and the default will never be used09:05
ogra_ToyKeeper, http://paste.ubuntu.com/10512940/ this one should work09:05
ogra_jibel, yes, fixed09:05
ToyKeeperOkay, I'm fine with a different method/patch...  so long as it will "just work" with no changes to files from the distro or tools which rely on phablet-network.  (so, behavior specified by env var or checking for a special file)09:06
ogra_ToyKeeper, yeah, no prob with changing the coe ... i just found your change a bit huge for the task :) test my change, if it works for your usecase we can land it09:07
ogra_*code09:07
* popey returns09:14
jibelToyKeeper, maybe using the network file provided on the command line, then ~/.phablet-network, then the system configuration then fail would do what you want. It won't change phablet-network's behaviour for current users, no change to the argument, no env to specify, and minimal changes to phablet-network and fixes the problem for automated tests running on machines without NM or without a Wifi09:16
jibelconnection.09:16
ToyKeeperjibel: Yup, that's what I was just doing.  :)09:17
* jibel re-reads your patch :)09:18
ToyKeeperI mean, what I was just redoing; no point having two slightly different formats for the wifi auth file.09:19
ogra_jibel, that is what my patch does ...09:21
ogra_well, in a different order ... if ~/.phablet-network exists it gets used, else it uses NMs default file unless there is -n09:22
ogra_http://paste.ubuntu.com/10512940/09:22
ToyKeeperogra_: Thanks.  Mostly, I wanted to make sure the idea was okay, and that someone on the project was aware of the patch.  :)09:25
ogra_thanks for bringing it up :)09:26
ToyKeeperogra_: New patch is uploaded, much simpler than the first, and does as jibel described.09:26
ogra_one small nitpick, can you put the actual filename in the usage line too ?09:27
ogra_(or uses NM-format wifi config from $DEFAULT_NETWORK_FILE which defaults to ...)09:27
ToyKeeperogra_: It *is* in the usage line.  Did you try running it?09:27
ogra_i dont see usage pointing to ~/.phablet-network in your code09:28
ogra_only to the var ... it should tell you that there is a default value for the var09:29
ToyKeeperThe variables get expanded when it prints the help string.09:29
* ogra_ slaps forehead09:29
ogra_sorry ...09:29
ogra_i need more coffee :)09:29
ogra_ToyKeeper, approved09:30
ToyKeeperI can't say I'm fully awake either...09:33
ogra_:)09:36
rvrogra_: ping10:41
ogra_rvr, hey, so what failed exactly ?10:41
rvrogra_: push/pull10:41
rvrogra_: They work when the screen is locked10:41
ogra_hmm, did you reboot after installing the package ?10:42
rvrYes10:42
rvrBut let me recheck again10:42
rvrAfter reboot10:44
rvr$ adb push diff.diff /home/phablet10:44
rvr645 KB/s (26568 bytes in 0.040s)10:44
ogra_hmm, did you ever install that phone with --developer-mode ?10:44
ogra_check if there is a dev-mode override file in place, it is called /userdata/.adb_onlock10:45
ogra_that would disable the lock screen check altogether10:45
rvrogra_: Let me see10:48
rvrogra_: Yes, the file exists10:49
ogra_remove it and reboot10:49
ogra_so we know at lest the behavior is correct with the file in place :)10:49
rvr:D10:49
ogra_i should have added that to the test plan, sorry10:50
rvr$ adb push diff.diff /home/phablet10:51
rvrerror: closed10:51
ogra_perfect :)10:51
rvrPerfect10:51
ogra_(sadly the error message is hardcoded on the PC side ... i would have liked something like "error: screen locked" )10:51
rvrThe other tests passed, I'm approving the silo10:52
ogra_thanks :)10:52
bzoltan_Mirv: we have started the final Vivid landing ... I hope it will make it11:08
sil2100bzoltan_: hey! The final big release of UITK to vivid - does that have any features?11:08
* sil2100 assigned a silos11:08
sil2100*silo even11:08
sil2100bzoltan_: but if it's feature-packed, then we'll probably have to file an FFe for it11:08
sil2100FFe's are the cool thing this cycle11:09
bzoltan_sil2100:  at this phase all new UITK festures address long pending bugs11:09
bzoltan_sil2100:  it is a big one, packed with 13 bugfixes. Who's call is to make an FFe or not?11:12
sil2100It's usually the release team deciding, as if we publish it'll require their approval - if they find that it requires an FFe, it needs to be filled in, but it's always best to try knowing that beforehand11:13
sil2100Let's ask the release team, sometimes they give us FFe's quite easily for typical touch/next components11:14
sil2100bzoltan_: is the UITK used anywhere in the desktop, besides Desktop Next?11:14
bzoltan_sil2100: our own QtCrator does, but not the version this landing has.11:15
bzoltan_sil2100:  no other application depends on it11:15
Mirvbzoltan_: good luck!11:17
bzoltan_Mirv: sil2100: i have added the gles branch too. Would you please reconf teh silo?11:23
sil2100bzoltan_: sure ,on it11:24
=== MacSlow is now known as MacSlow|lunch
Mirvbzoltan_: you'll need to bzr rm debian/patches in your -gles branch, since the real fix is in staging now and conflicts12:00
bzoltan_Mirv:  OK, thanks12:05
davmor2ogra_, sil2100: so it appears there is a custom-tarball option in ubuntu-device-flash it just doesn't work12:05
* sil2100 off to lunch12:05
ogra_davmor2, time for a bug then12:05
davmor2ogra_: indeed12:06
=== chihchun is now known as chihchun_afk
jibeldavmor2, I filed bug 142766712:46
ubot5bug 1427667 in goget-ubuntu-touch (Ubuntu) "option --custom-tarball doesn't work" [Undecided,New] https://launchpad.net/bugs/142766712:46
davmor2jibel: nice12:47
=== MacSlow|lunch is now known as MacSlow
alexabreuMirv, ack (for silo 13)13:32
=== chihchun_afk is now known as chihchun
ogra_imgbot, status 119 vivid14:02
imgbotStatus: succeeded, Started: 2015-03-03 02:02:09 UTC, Finished: 2015-03-03 02:56:07 UTC14:03
imgbotBuild URL: https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu/vivid/ubuntu-touch/+build/2166314:03
imgbotChangelog: http://people.canonical.com/~ogra/touch-image-stats/119.changes14:03
ogra_sil2100, rsalveti, once the livecd-rootfs upload i just did lands in the archive, i want to do a re-build (need the log info)14:27
ogra_for vivid that is14:28
rsalvetisure14:30
kenvandinedbarth, what's the status of rtm silo 10?14:35
kenvandinedbarth, i'd like to prepare another landing for settings soon14:36
popeyballoons: what's happening with https://code.launchpad.net/~nskaggs/ubuntu-calendar-app/fix-infloop-ap-trunk/+merge/251122 ?14:46
popeyballoons: It appears to be blocking https://code.launchpad.net/~pkunal-parmar/ubuntu-calendar-app/LiveEventModification/+merge/24771114:47
popey(which I am gettng increasingly worried about)14:47
* balloons erases his original comment14:51
balloonspopey, specific to that merge, I created it to help kunal's merge and fixup the tests. I filed several more bugs as well around calendar as you know. It seems trunk is not stable14:52
sil2100ogra_: sure thing14:53
balloonspopey, I was also going to mention calendar seems to have struggled in the past, but we've done work to make it well tested and working. It seems to break down over time14:54
balloonsI can only assume the previous tests were not completely stable and they are changing code and not tests as time goes on14:55
Mirvsil2100: 5.4.1 not all sun and flowers, so no early signoff testing this week at least. no other problems seen yet but there's a problem with the keyboard for unknown reason.14:55
Mirvsil2100: if there are any big issues, it would mean I'd land it to vivid after the fork, not before...14:56
ogra_if that publisher will ever move ...14:56
sil2100Mirv: ouch, you mean the OSK doesn't work properly?14:56
balloonspopey, the weird part is how trunk manages to get broken tests in it. This whole 'run it again' when jenkins doesn't pass it is how that happens.14:56
popeyballoons: I landed a few merges a couple of weeks back and had to retry a few times14:57
Mirvsil2100: you could say so... a crash bug #1427710 - tsdgeos might have time to look at it at some point. but it better have a good explanation (plausible, there is always some room for something not noticed) or otherwise Qt 5.4.1 starts to sound too risky.14:57
popeyballoons: this one just refuses to pass14:57
ubot5bug 1427710 in maliit-framework (Ubuntu) "A maliit-framework segfault with Qt 5.4.1" [Undecided,New] https://launchpad.net/bugs/142771014:57
Mirvsil2100: during the meeting I hadn't yet tried to _type_ anything, just playing media and swiping along :)14:58
balloonspopey, right.. I've re-run things as well, but doing that rather than fixing the issue does lead to messes like this14:58
sil2100hah ;)14:58
popeyballoons: is there someone who can help us clean these up?14:58
tsdgeosMirv: sil2100: need to talk to Saviq later to see if i can sneak some time into this14:59
balloonspopey, I mention that they have struggled because I've seen things removed / commented out in the tests when I went to correct the loop issue in my mp15:00
Mirvsil2100: meanwhile, we do have FFe granted...15:00
sil2100Mirv: oh!15:00
popeyballoons: we need a way forward, because right now we're stuck and nothing is landing15:00
balloonspopey, yes, I noticed Carla actually has been submitting calendar mp's before I even asked her or Daniel to look15:00
balloonspopey, this is marked critical, so in theory it should be first on the list for folks: https://bugs.launchpad.net/ubuntu-calendar-app/+bug/142618315:03
ubot5Launchpad bug 1426183 in Ubuntu Calendar App "rev593 (trunk) autopilot tests experience random failures" [Critical,Triaged]15:03
popey:(15:04
balloonsaww, now I feel sad15:05
popeywelcome to my world15:06
dbarthkenvandine: you could take the branch if you want; i'd still like to get dobey's approval on that one though15:26
dbarthbrb15:26
dobeywhat's up?15:26
kenvandinedbarth, i'll leave it to you, i mostly wanted to know when you thought it would get an ack... so i can plan when i can do my landing or if i should try to get in front of you in line15:27
kenvandinedobey, rtm silo 1015:27
dobeywell the branch is already approved15:27
dobeythere is some other issue in rtm that makes it sort of not work right though15:28
kenvandinei'm asking about the silo15:28
dobeyor well, i'm not sure if it's another issue in rtm, or a feature15:28
dobeythe code is correct though15:29
dbarthdobey: so the hook that cleans up the account acl is fine to land ?15:47
dobeywait, what?15:48
dobeyno, i don't think we should land that15:49
dobeywhy is that in there?15:50
dobeythat code doesn't exist in vivid, so i don't think we should land it in rtm. and we don't fully understand what is happening on boot in rtm, so landing a "fix" without understanding the issue seems wrong to me.15:52
imgbot=== IMAGE 120 building (started: 20150303-15:55) ===15:55
dobeymardy: won't this hook fail anyway, because it's unconfined and unconfined isn't in the acl, so it can't actually do queryInfo() on the identity?15:58
dobeywell i need to get lunch16:20
ogra_jibel, could you try image 120 (once it is done) wrt /var/log ownership ... according to the buld log it is definitely syslog owned now16:26
jibelogra_, what is 120? 130 is in devel-proposed for krillin16:27
ogra_jibel, ah, the both goes after mako versions :)16:27
ogra_*bot16:27
jibelok16:27
ogra_jibel, the image that is just building ... should be done soon16:28
jibelI'll try whatever is in proposed16:28
Saviqtrainguards, can you please remove qtmir{,-gles} from vivid silo 1916:47
sil2100Saviq: sure, from the PPA you mean?16:48
Saviqsil2100, yes, because the build job lied to me again :P16:48
* Saviq tries to find the bug16:48
sil2100Oh, so it seems it still doesn't remove packages when reconfiguring ;/16:48
Saviqyeah16:49
sil2100Saviq: should be gone from the PPA now16:51
Saviqom26er, ↑16:51
Saviqsil2100, thanks16:51
om26erSaviq, great, thanks16:52
bzoltan_sil2100:  my krillin dived into a reboot loop and even the recovery mode does not bring it out17:17
bzoltan_ogra_: ^17:18
bzoltan_or anybody :(17:18
imgbot=== IMAGE 120 DONE (finished: 20150303-17:20) ===17:20
imgbot=== changelog: http://people.canonical.com/~ogra/touch-image-stats/120.changes ===17:20
sil2100bzoltan_: huh?17:20
sil2100bzoltan_: how? What did you do to get it in this state?17:20
bzoltan_sil2100:  the AP tests of the dialer app ... it does it all the time17:22
bzoltan_sil2100:  but this time I can not do anything with it17:22
sil2100ogra_: ^ do you know what can be done in this case?17:23
ogra_nope17:23
ogra_well, first of all file a bug against that ap test17:23
ogra_then try to get into fastboot mode and flash the open recovery.img17:23
ogra_from there you should be able to do u-d-f17:23
bzoltan_ogra_: sil2100: anything what could bring back this device to a working state?17:23
bzoltan_ogra_:  how to flash the open recovery.img?17:24
ogra_bzoltan_, by following the instructions from the mailing list17:25
bzoltan_ogra_:  with what subject?17:25
* bzoltan_ gets hundreds of mails on many ML17:25
ogra_something about adb and recovery17:25
bzoltan_ogra_:  I am not much closer...17:27
ogra_popey, ok, all links restored for rtm17:30
popeyyay17:31
popeythanks17:31
bzoltan_ogra_:  sil2100: and where to get the recovery.img from?17:35
ogra_bzoltan_, click the link in the mail17:35
ogra_it has all instructions17:35
jibelogra_, on 131 /var/log is owned by root:syslog17:49
ogra_\o/17:49
jibelwhat changed?17:50
ogra_the code used "chgrp syslog /var/log"17:50
ogra_not sure why that didnt work17:50
ogra_i switched it to"chown root:systlog /var/log"17:50
ogra_(without the typo :P )17:50
ogra_so we should be fine now ...17:51
ogra_what scares me a bit is that there could potentially be other dirs we dont cover like this17:51
ogra_syslog is just sticking out because we all look at it all the time17:51
jibelindeed that was my concern with this bug17:55
ogra_right, but not much we can do except keepin our eyes open17:55
bzoltan_ogra_:  I did not find "the" mail .. I have 11k mails in that folder and I did not find the one with instructions on how to flash the recovery.img17:57
ogra_bzoltan_, i get between 600-800 mails per day, dont try to get into a pissing contest with me :P17:58
popeybzoltan_: "adb no longer available in recovery on krillin/vivid (and rtm too)"17:58
popeythat's the subject17:58
bzoltan_ogra_:  you win me by far great master :) but that does not help my device17:59
ogra_bzoltan_, well, what popey said17:59
* popey wins at making the google mail bots search his mail17:59
ogra_lol, i wont give them access to my imap server :)17:59
bzoltan_ogra_:  how tricky .. it was on the phablet ML18:00
popeyya18:00
ogra_(i'm pretty sure i would have found it in minutes if i had the time for searching)18:00
davmor2bzoltan_, ogra_: Can I join in the pissing contest I get around 800-1200 emails a day when I take a week off it takes 3 weeks to get fully caught up again. And even I know about the adb not working any more mail.  Also I have several folders around the 20-30000 email mark and 3 over 50,000 and I haven't been here as long as ogra_ so I assume his folders are bigger still :D18:03
ogra_my whole /var/mail/ogra on the server contains around 5mio mails18:04
ogra_dating back to before i started at canonical18:04
bzoltan_davmor2: ogra_: with 3years I am junior here ... but still I am surprised that ML archive here is considered as documentation storage :)18:04
davmor2ogra_: yeah but your old dude ;)18:04
* ogra_ shakes his cane18:05
ogra_get off my lawn !18:05
davmor2can't you see the sign18:05
ogra_:D18:05
ogra_bzoltan_, well, it helped me a lot to have it over the last decade18:06
ogra_and evolution is reasonable fast at searching piles of mail18:06
ogra_even at that size18:06
davmor2ogra_: claws is blistering quick18:06
bzoltan_ogra_:  searching in mails my client is fast too.. once I figure out the right folder :)18:07
davmor2bzoltan_: just search them all ;)18:07
* bzoltan_ checks the oldes mail ... yes it came in 9618:07
dobeysearching in /dev/null is the fastest evar18:08
ogra_dobey, yeah, but the results are so unsatisfying18:08
dobeys/un// :)18:08
ogra_haha18:08
dobeythe oldest mail in my inbox is from "?"18:09
dobeyyay spam18:09
ogra_lol18:09
dobeyoldest actual mail i have is i think ~1997ish18:12
Laneyman I wasn't even born then18:12
bzoltan_mine is from a mailing list18:12
davmor2bzoltan_, ogra_: my first mail on my home server is 2006 and in gmail is 2004 -ish18:13
davmor2Laney: bigRon at our lug can tell you about the days they invented the interwebz ;)18:14
bzoltan_davmor2:  I am confused with my records... I wrote a mail server and client in 92 for novel netware 3.1 without nowing about the existence of internet/email do those mails count?18:16
* bzoltan_ gets sensitive about the golden past18:16
=== alan_g is now known as alan_g|EOD
davmor2bzoltan_: hahaha18:42
bzoltan_ogra_:  fastboot - u-d-f - reboot cycle again ...18:51
bzoltan_does anybody know what the hack I do wrog here https://ci-train.ubuntu.com/job/ubuntu-landing-003-1-build/147/console ?20:26
robrubzoltan_: yeah, -gles isn't configured in the silo21:06
bzoltan_robru: do we need tro reconf?21:28
robrubzoltan_: yeah, seems so. one sec21:29
robrubzoltan_: ok try now21:30
robrutedg23:30
robrunuddy23:31
robrubuddy23:31
robrulol23:31
tedgrobru, I'm getting ready to head out, is that a purposeful ping? :-)23:31
robrutedg: yes. you apparently didn't manage your silo conflicts very well23:31
tedg?23:31
robrutedg: there it is ^23:31
robrutedg: you had url-dispatcher in silo 4 and 823:32
tedgHmm... I don't have an RTM silo.23:32
robrutedg: you published 8 and then didn't rebuild 4.23:32
robrutedg: but you do? http://people.canonical.com/~platform/citrain_dashboard/#?distro=ubuntu-rtm&q=tedg23:32
tedgrobru, That branch landed, someone must have reallocated the silo after I deallocated it.23:32
tedgPerhaps a race condition23:33
tedg(human)23:33
tedgLet me kill it.23:33
robrutedg: weird, it was built by bfiller: https://ci-train.ubuntu.com/job/ubuntu-rtm-landing-004-1-build/104/console23:33
robrutedg: ok well if you're sure it's wrong, you can kill it. also the mp is merged, so that's curious.23:34
tedgI marked it as not ready anymore.23:34
robrutedg: thanks23:34
tedgrobru, The MP was in silo8, it landed with the webbrowser changes.23:34
tedgI originally had a silo for the MP so that the browser folks could test, but then they just worked on vivid since it was there already.23:35
tedgAnyway. Useless history :-)23:35
tedgrobru, Thanks for mentioning it, cleaned it up :-)23:35
tedg'night folks.23:35
robrutedg: night23:35

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!